section 4 : measurement and analysis of agroforestry experiments

Recent developments in homoclime analysis to assist species selection

T. H. Booth

CSIRO Division of Forest Research
 P.O. Box4008, Canberra A.C.T. 2600, Australia

Abstract

Two methods of homoclime analysis have been devised to assist the selection of Australian tree species for trials. Both methods use modem interpolation techniques to estimate meteorological variables at sites which may be some distance from recording stations. The first requires no biological information, but simply compares climatic conditions at a trial site outside Australia with conditions of 2795 locations in a regular half-degree grid across Australia. The second uses information from a species' natural distribution and results of its performance in trials to develop a detailed description of its climatic requirements.

Introduction

The Australian Centre for International Agricultural Research (A C I A R) is supporting studies of the potential of Australian tree species for fuelwood and agroforestry use in developing countries (Turnbull 1986).

Two methods of homoclime analysis have been developed to assist the selection of species for trials. Both methods make use of modern interpolation methods, which provide better estimates of mean climatic conditions for remote sites than techniques previously used. For example, the surfaces fitted for Australia estimate mean monthly values of average daily maximum and minimum temperature with mean errors of 1.3 and 4.1 per cent respectively. Errors associated with estimates of monthly mean precipitation are generally below 10 per cent. The interpolation methods used were devised by Wahba and Wendelberger (1980) and implemented by Hutchinson (in press), using data collected by Nix and McMahon (McMahon 1986, p. 60-63).

Methods

The first method, known as CLIMSIM (climatic similarity), allows any location in the world to be analysed, provided monthly mean values of daily maximum temperature, daily minimum temperature and precipitation are available. From the 36 monthly values, 18 climatic factors, such as annual mean temperature and driest quarter precipitation, are calculated. These are compared with the same 18 factors estimated at 2795 locations in a half-degree grid across Australia. A map is plotted which indicates the locations in Australia which have conditions most similar to the target site (Figure 1). Booth et al. (1987) describe the method in detail and use examples from Africa and South America to show how the method can assist tree species selection.

Figure 1 Locations in Africa which satisfy three main climatic characteristics derived from the natural distribution of A. auriculiformis in Australia.

The second method is an extension of the BIOCLIM (bioclimatic prediction) system devised by Nix, Busby and Hutchinson (see Nix 1986, p. 415; Busby 1986; Hutchinson in press). Booth (1985) used the BIOCLIM program to analyse climatic variability in the natural distribution of Eucalyptus citriodora. This information was then used to identify homoclime sites in Africa, which were shown to correspond to many of the areas where E. citriodora had proved successful in trials.

Booth et al. (in press) analysed the natural distribution of a further twelve eucalypt species, but also showed how results from trials in Africa could be used to improve estimates of the species' climatic requirements. The BIOCLIM program has recently been used to analyse climatic variability in the natural distribution of ten non-eucalypt species which have potential for fuelwood and agroforestry.

Comment

This brief note is intended to draw the attention of researchers to these recent developments. Further details, including an outline of both methods (Booth in press), are available on request.

Although the examples cited above have been carried out using Australian species there is no reason why the same methods could not be used to analyse the requirements of other species, provided the appropriate surfaces were developed for climatic interpolation.

Acknowledgements

I am grateful to The Australian Centre for International Agricultural Research (ACIAR) for financial support; Henry Nix (Centre for Resource and Environmental Studies, Australian National University); Michael Hutchinson (CSIRO Division of Water and Land Resources); and to June McMahon (CSIRO Division of Water and Land Resources) for access to programs and data.

References

Booth, T.H. 1985. A new method to assist species selection. Commonw. For. Rev. 64: 241-50.

Booth, T.H. (in press). Which wattle where?: selecting Australian acacias for fuel-wood plantations. Plants Today

Booth, T.H., H. A. Nix, M. F. Hutchinson and J. R. Busby, 1987. Grid matching: a new method for homoclime analysis. Agric. For. Meteorol. 39(2-3): 241-255.

Booth, T.H., H. A. Nix, M. F. Hutchinson and T. Jovanovic, (in press). Niche analysis and tree species selection. For. Ecol. Manage.

Busby, J.R.1986. A biogeoclimatic analysis of Nothofagus cunninghamii (Hook.) Oerst. in south eastern Australia. Aust. J. Ecol. 11:1-7.

Hutchinson, M.F. (in press).A new objective method for spatial interpolation of meteorological variables from irregular networks applied to the estimation of monthly mean solar radiation, temperature, precipitation and windrun. In Need for climatic and hydrologic data in agriculture of south-east Asia. Tokyo: United Nations University.

McMahon, J.P. 1986. Climatic databases held in the Division. Research Report 1983-85. Canberra: CSIRO Division of Water and Land Resources.

Nix, H.A. 1986. A biogeographic analysis of Australian elapid snakes. In R. Longmore (editor),Atlas of Australian elapid snakes. Canberra: Bureau of Flora and Fauna.

Turnbull, J.W. (ed). 1986. Multipurpose Australian trees and shrubs. Canberra: Australian Centre for International Agricultural Research. 316 pp.

Wahba, G., and J. Wendelberger. 1980. Some new mathematical methods for variational objective analysis using splines and cross validation. Mon. Weather Rev. 108:1122-43.